THE EVOLUTION OF STELLAR SYSTEMS 83 



alcohol and water of a specific gravity as nearly as possible 

 equal to that of oil. Some oil was then poured into the mixture. 

 As the bottom of the mixture was slightly more dense than 

 the oil and the top slightly less dense, the oil sank half-way 

 and floated in the middle as a round ball. By means of a disc 

 attached to a wire the ball of oil was set rotating. The effect 

 of the rotation caused the oil globe to expand into the form 

 of a spheroid flattened at the poles, and this flattening increased 

 with. the speed, until at last a ring was formed which revolved 

 round the globe. After a time the ring broke up and gathered 

 into a smaller globe which rotated, besides revolving round 

 the large globe. 



Laplace supposed that the rings would rotate as though 

 solid, their outer edges thus moving more swiftly than the 

 inner, and thus the planets formed therefrom would rotate in 

 the same direction. The " exceptional cases " of our system — 

 the fact that the satellites of Uranus and Neptune move in the 

 opposite direction to that in which most of the other members 

 do, and the swift revolution of the inner satellite of Mars— cannot 

 be explained by this form of the hypothesis. 



M. Faye, however, by modifying the original idea of Laplace 

 and supposing that the planets were formed by local condensa- 

 tions (not by the detachment of rings) within the revolving 

 nebula, and that the outer planets, Uranus and Neptune, have 

 been more recently formed than the rest, has shown that these 

 bodies would have retrograde rotation on their axes, which he 

 supposed to be the case from the motion of their satellites. 

 Since, however, Saturn's rotation is in the same direction as 

 that of our own earth, and eight of its satellites move in one 

 direction, but the last discovered (Phoebe) moves in the opposite 

 direction, we have still a difficulty, unless we suppose this body 

 to be a recent capture and not an original member of the Saturnian 

 family. The same thing is the case also with the eighth satellite 

 of Jupiter, whose motion is retrograde, whilst the other seven 

 have direct motion. Prof. Sir George Darwin by his theory 

 of Tidal Evolution has attempted to explain the swift motion 

 of the inner satellite of Mars, the fact that the moon always 

 turns the same face towards the earth, and that the rotation 

 period of Mercury (and probably that of some of the satellites 

 of Jupiter and Saturn) is the same as that of its revolution. 

 He has given reasons for thinking that in former ages the period 



